Tesla M2070 and GeForce RTX 4070 SUPER - video card comparison

Tesla M2070

GeForce RTX 4070 SUPER

Architecture

Fermi (2010−2014)

Ada Lovelace

GPU code name

GF100

AD104

Market segment

Workstation

Desktop

Release date

July 2011

January 2024

Interface

PCIe 2.0 x16

PCIe 4.0 x16

Supplementary power connectors

1x 6-pin + 1x 8-pin

1x 16-pin

Boost clock speed

2475 MHz

Core clock speed

574 MHz

1980 MHz

Floating-point performance

1,030.4 gflops

Manufacturing process technology

40 nm

5 nm

Number of transistors

3,100 million

35,800 million

Pipelines / CUDA cores

448

7168

Power consumption (TDP)

225 Watt

220 Watt

Texture fill rate

32.14

584.6

Memory type

GDDR5

GDDR6X

Maximum RAM amount

6 GB

12 GB

Memory bus width

384 Bit

192 Bit

Memory clock speed

3132 MHz

21000 MHz

Memory bandwidth

150.3 GB/s

504.2 GB/s

Shared memory

Display Connectors

No outputs

1x HDMI 2.1, 3x DisplayPort 1.4a

HDMI

DirectX

12 (11_0)

12 Ultimate (12_2)

CUDA

2.0

8.9

OpenCL

1.1

3.0

OpenGL

4.6

Shader Model

5.1

6.7

Vulkan

N/A

1.3